Many persons with cancer hold beliefs that are barriers to optimal pain management, such as exaggerated fears of addiction and concerns about medication tolerance (Ferrell, et al., 1993; Ward, et al., 1993). Although educational interventions have been shown to overcome these patient-related barriers and improve pain management, they have not been widely disseminated and relatively few persons with cancer have had the benefit of systematic education about pain management. The proposed study will establish a foundation for addressing this problem. The specific aims are to test the efficacy and the cost-effectiveness of a Tailored Barriers Intervention (TBI), an intervention consisting of tailored proactive educational messages. After receiving usual service, patients calling the Cancer Information Service (CIS) will be invited to join a study. Those agreeing to join and who have moderate to severe pain will be randomized to one of three groups. Group 1 is a no-assessment no-intervention control group. Group 2 receives baseline assessment but no intervention (to control for possibility that assessment serves as an intervention). Group 3 receives baseline assessment plus TBI. Subjects in all three groups will be telephoned by the UW Survey Research Shared Service (SRSS) and asked to complete measures of barriers, pain duration, pain severity and pain interference with life activities 28 days after entering the study. This design assures that study findings will be optimally useful in planning the next steps in dissemination. If the trial reveals that TBI is efficacious and cost-effective, then a next study will be planned to investigate the feasibility of instituting TBI as part of the usual care provided by CIS staff to callers. If, on the other hand, the baseline assessment alone is just as effective as the TBI, or if it is more cost-effective, then the next study would focus on the feasibility of instituting barriers assessment as part of usual care provided to CIS callers.